Mycobacterium tuberculosis kills more people than any other single infectious agent. In spite of the continuing global health threat of tuberculosis and the prevalence of drug-resistance tuberculosis in the United States, new strategies for the diagnosis, prevention and cure of tuberculosis have been slow to emerge. Considerable advances have been made in mycobacterial genetics in recent years, and the development of methods for transformation, mutation, and construction of recombinant strains has facilitated more sophisticated genetic analyses. However, the construction of defined gene replacements and null mutants in M. tuberculosis continues to be problematic, requiring multi-step plasmid or phage constructions to make the desired mutants. Moreover, these multi-step processes often take many weeks or months in M. tuberculosis due to its extremely slow-growth with a 24 hours double-time. Genetic methods for recombineering have been developed for E. coli and its genetic elements that take advantage of high-frequency phage-mediated recombination systems to construct mutants in a single step, using either PCR products or oligonucleotides. Adaptation of the these E. coli systems to the mycobacteria has yet to be successful, suggesting that a more fruitful approach may be to develop recombineering systems based on mycobacteriophage-encoded recombination systems. The development of a mycobacterial recombineering system would enable the construction of a defined set of genetic disruptions covering the entire M. tuberculosis genome that would provide a useful recourse to the entire community of mycobacterial geneticists. Preliminary data support the hypothesis that mycobacteriophage TM4 encodes its own recombination functions, since combinations of TM4 cosmids give rise to wild-type TM4 plaques when co-electroporated into M. smegmatis. Moreover, the formation of these recombinants is independent on either the host RecA or RecB activities. Since TM4 does not encode any recognizable recombination genes, it is plausible that TM4 encodes a novel recombination system that can be exploited for recombineering purposes. Moreover, both mycobacteriophage Che9c and Halo contain genes that do encode either RecE- or RecT-like proteins that are predicted to mediate phage recombination systems. These putative recombination proteins will be functionally and biochemically characterized and used to develop recombineering systems in both fast- and slow-growing mycobacteria.